2pm Pacific Time
BlueJeans: https://bluejeans.com/426716450
attending:
regrets:
- John Swinbank (review)
Topics for discussion:
- sample alerts for brokers (Eric Bellm , Unknown User (cmorrison) , Spencer Nelson ):
- alert sizes: sample alerts are too big for the stream simulator. Long-term fix is to work on cutouts, etc. Short term, just filter the sample?
- current limit is 10 MB
- Spencer worries that if we really are going to have messages that are super big
- alert sizes: sample alerts are too big for the stream simulator. Long-term fix is to work on cutouts, etc. Short term, just filter the sample?
- Gen 3 conversion (all):
- Krzysztof Findeisen:
- working on gen-3 ap_verify: ap_pipe + metrics piplines
- how do we want to organize the Metrics?
- don't want one pipeline per metric; but may need more granularity than "compute & store all metrics"
- runtime/memory are already computed by ap_pipe, metric task is just gathering these from task metadata
- is there a functional separation? quantities that are "easy" in terms of IO/computation vs. "hard"? This may be too obscure to maintain though
- Chris M wonders about bundling by PipelineTask (ie, ip_diffim). Eric: or maybe the data products they touch? source catalogs, apdb...
- Meredith: leave it all on for now and wait until we need to think about it? KF worries people need an example of multiple pipelines so they don't simply continue appending to a single giant file
- KPMs or requirements could be pulled out separately?
- consensus:
- runtime
- Precursor processing (Meredith Rawls ):
- question for NCSA: how to we run on something other than SQLite?
-
DM-25989Getting issue details...
STATUS
- Meredith Rawls will ping Unknown User (emorganson) on ap channel
- finished July reprocessing
- spawned a bunch of HSC tickets for cleanup
- question for NCSA: how to we run on something other than SQLite?
-
DM-25989Getting issue details...
STATUS
- report from NCSA (Unknown User (emorganson) + Unknown User (manninga) et al):
- out today
- Alert Distribution (Spencer Nelson ):
- working on presentation for split alerts for SST: Eric sent estimate of a minimal ~100 byte/alert that still enables some initial triage
ap_
association/SDM/fakes/alerts (Unknown User (cmorrison)):- merge to limit cutout sizes is in:
-
DM-25930Getting issue details...
STATUS
- Eric Bellmneeds to check with Solar Sys folks about max length of NEO streaks
- Siegfried Eggl reports "I was looking at NEO Duende during its close approach in 2013 (within 40000km of the Earth) and for a 30s exposure I get roughly 0.4-0.5 deg in streak size on sky."
- working on detection footprint questions
- back to fakes, iterating with Sophie on a review
- thinks should be straightforward to convert ap_association to use MySQL; bug dax if not
- merge to limit cutout sizes is in:
-
DM-25930Getting issue details...
STATUS
- Crowded field photometry (Ian Sullivan ):
- fighting NOAO archive server–got blocked, can't get support
- Eric suggests pinging John and I to find the appropriate contact in Tucson
- otherwise debugging ap_pipe run
- fighting NOAO archive server–got blocked, can't get support
- Image differencing (Unknown User (gkovacs) ):
- out today
- Review CI (https://chronograf-demo.lsst.codes/):
- Review outstanding action items
- AOB
Action Items
Description | Due date | Assignee | Task appears on |
---|---|---|---|
| 17 Jul 2023 | Eric Bellm | AP Pipeline Meeting, 2023-06-26 |
| 11 Sep 2023 | Eric Bellm | AP Pipeline Meeting, 2023-08-21 |
| 25 Sep 2023 | Eric Bellm | AP Pipeline Meeting, 2023-08-14 |